1. Field of the Invention
The present invention relates to image processing, and more particularly, to an image processing circuit and a method thereof.
2. Description of the Prior Art
According to characteristics of human eyes and some assumptions related to photoreception of human eyes, color information recorded in many image compression formats is less than brightness information therein, in order to improve the compression rate of video data with acceptable image quality to human eyes, so the storage or transmission efficiency thereof can be improved. For example, in a video processing procedure of a Digital Versatile Disc (DVD) player, transformations between any two of so-called “4:4:4” format, “4:2:2” format, and “4:2:0” format are involved. In the three formats mentioned above, the “4:4:4” format includes the most color information, and the “4:2:0” format includes the least color information. Therefore, the “4:2:0” format provides the best compression rate in contrast to the other two formats.
The “4:4:4” format mentioned above means: for every four brightness samples Y, there exists four color samples Cb and four color samples Cr corresponding to the four brightness samples Y. In addition, the “4:2:2” format means: for every four brightness samples Y, there exists two color samples Cb and two color samples Cr corresponding to the four brightness samples Y. It is noted that the number of the color samples Cb and the number of the color samples Cr along a horizontal direction of an image frame in the “4:2:2” format are respectively reduced to half of those in the “4:4:4” format. Additionally, the number of the color samples Cb and the number of the color samples Cr in the “4:2:0” format are respectively reduced to half of those in the “4:4:4” format along either a horizontal direction or a vertical direction of an image frame. The relationships between the three formats mentioned above are well known in the art and therefore not explained in detail here.
When an MPEG decoder of the DVD player converts video data of the “4:2:0” format to video data of the “4:2:2” format, additional color samples Cb and Cr should be generated along a vertical direction of an image frame. If the MPEG decoder utilizes a field-decoding method to perform the format conversion mentioned above, errors would easily occur considering a frame composed of an odd field and an even field generated by this format conversion. Please refer to FIG. 1. FIG. 1 illustrates working flows of format conversions respectively utilizing a duplication procedure 120 and an interpolation procedure 140 of the field-decoding method according to the prior art, where the source image 110 can be converted into the duplicated image 130 and the interpolated image 150 respectively through the duplication procedure 120 and the interpolation procedure 140. In FIG. 1, only two pixels are illustrated in each image for the horizontal direction, where different patterns (with different line densities) in these pixels of the images represent different colors. Please note that common boundaries 112 and 114 of objects in the source image 110 are distorted or falsified after the duplication procedure 120 or the interpolation procedure 140, as shown by the duplicated image 130 and the interpolated image 150.